A biological age model based on physical examination data to predict mortality in a Chinese population
Qingqing Jia,
Chen Chen,
Andi Xu,
Sicong Wang,
Xiaojie He,
Guoli Shen,
Yihong Luo,
Huakang Tu,
Ting Sun,
Xifeng Wu
Affiliations
Qingqing Jia
Department of Big Data in Health Science School of Public Health, Center of Clinical Big Data and Analytics of The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310058, China
Chen Chen
Department of Big Data in Health Science School of Public Health, Center of Clinical Big Data and Analytics of The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310058, China
Andi Xu
Department of Big Data in Health Science School of Public Health, Center of Clinical Big Data and Analytics of The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310058, China
Sicong Wang
Department of Big Data in Health Science School of Public Health, Center of Clinical Big Data and Analytics of The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310058, China
Xiaojie He
Health Management Center, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou 310009, China
Guoli Shen
Health Management Center, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou 310009, China
Yihong Luo
Department of Big Data in Health Science School of Public Health, Center of Clinical Big Data and Analytics of The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310058, China
Huakang Tu
Department of Big Data in Health Science School of Public Health, Center of Clinical Big Data and Analytics of The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310058, China
Ting Sun
Health Management Center, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou 310009, China; Corresponding author
Xifeng Wu
Department of Big Data in Health Science School of Public Health, Center of Clinical Big Data and Analytics of The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310058, China; National Institute for Data Science in Health and Medicine, Zhejiang University, Hangzhou, Zhejiang, China; The Key Laboratory of Intelligent Preventive Medicine of Zhejiang Province, Hangzhou, Zhejiang, China; Cancer Center, Zhejiang University, Hangzhou, Zhejiang, China; School of Medicine and Health Science, George Washington University, Washington, DC, USA; Corresponding author
Summary: Biological age could be reflective of an individual’s health status and aging degree. Limited estimations of biological aging based on physical examination data in the Chinese population have been developed to quantify the rate of aging. We developed and validated a novel aging measure (Balanced-AGE) based on readily available physical health examination data. In this study, a repeated sub-sampling approach was applied to address the data imbalance issue, and this approach significantly improved the performance of biological age (Balanced-AGE) in predicting all-cause mortality with a 10-year time-dependent AUC of 0.908 for all-cause mortality. This mortality prediction tool was found to be effective across different subgroups by age, sex, smoking, and alcohol consumption status. Additionally, this study revealed that individuals who were underweight, smokers, or drinkers had a higher extent of age acceleration. The Balanced-AGE may serve as an effective and generally applicable tool for health assessment and management among the elderly population.
